AI Strategy & Transformation
Connect AI investment with business priorities, operating-model transformation, governance and measurable value.
Explore advisory →DEVRAJ SARKAR · INDEPENDENT AI & CYBERSECURITY ADVISOR
AI Strategy, Governance & Cybersecurity for the Agentic Enterprise.
Advising C-suite and board leaders on the transition from AI experimentation to secure, governed and responsible autonomy.
Based in Kolkata, India, I advise organizations on building AI as a secure, governed and business-driven enterprise capability—bringing together AI strategy, Agentic AI, Responsible AI, AI governance, AI security, cyber risk and assurance.
23+Years in IT
15+Years Consulting
ISO/IEC 42001 & 27001Lead Auditor
C-Suite & BoardAdvisory
STRATEGY
AI
GOVERNANCE
SECURITY
TRUST
THE NEXT ENTERPRISE TRANSITION
AI is moving from answering questions to taking action.
Agentic AI systems can reason, plan, use tools, access enterprise data, collaborate with other agents and execute multi-step business processes. That changes the enterprise risk equation.
Understand the autonomy challenge →01ExperimentExplore possibilities
02AdoptScale use cases
03GovernEstablish controls
04DelegateEnable agents
05AutonomyOperate responsibly
ADVISORY FOCUS
Where AI, security, governance and business value converge.
Focused advisory and assurance for C-suite and board leaders moving from experimentation to enterprise-scale AI, in India and internationally.
Agentic AI Security
Design trusted autonomous systems with bounded authority, identity, tool controls and oversight.
Explore Agentic AI →Responsible AI & Governance
Build policy, accountability, risk classification, lifecycle controls and evidence.
Explore governance →AI Assurance
Assess whether governance structures and controls operate as intended and can be demonstrated with evidence.
Explore assurance →Cybersecurity & Digital Risk
Strengthen identity, cloud, secure engineering, Zero Trust, DevSecOps and digital risk governance.
Explore cybersecurity →ISO/IEC 42001 & AI Management Systems
Connect governance obligations, risk treatment, documented evidence, audit and continual improvement.
Explore ISO/IEC 42001 →FEATURED FOCUS · AGENTIC AI
The question is no longer just what AI can do. It is what AI should be allowed to do.
AI agents may determine how to achieve an objective, select tools, retrieve information, invoke APIs, collaborate with other agents and act on behalf of users or organizations.
What is AI Agent Security?
Explore Agentic AI security →
AI Agent Security is the discipline of protecting autonomous or semi-autonomous AI systems from excessive permissions, manipulated instructions, unsafe tool use, identity abuse, data leakage and unaccountable actions.
AGENT
AUTONOMY
IdentityWho is the agent?
AuthorizationWhat can it do?
MemoryWhat does it retain?
ToolsWhat can it invoke?
OversightWhen must humans approve?
AuditabilityCan actions be reconstructed?
RESPONSIBLE AUTONOMY MODEL
Autonomy is a spectrum. Governance should scale with it.
Assist
AI recommends. Human decides and acts.
Collaborate
AI performs defined tasks. Human supervises.
Delegate
AI executes approved processes within boundaries.
Autonomous
AI plans and executes with exception-based oversight.
Multi-Agent
Multiple agents coordinate decisions and actions.
THE AUDITOR'S PERSPECTIVE
Trust requires evidence.
AI governance cannot stop at policies and principles. Organizations need evidence that controls operate in practice.
Can an AI agent perform a high-impact action without independent authorization?
A manipulated or incorrectly configured agent could act beyond its intended authority.
Agent identities, IAM roles, API scopes, tool permissions, execution logs and approvals.
Explicitly scoped privileges and stronger approval for consequential actions.
RESEARCH
Analysis for people responsible for technology risk.
AI Agent Security
Identity, permissions, memory, tools, delegation and accountability in autonomous AI systems.
Research library →ISO/IEC 42001
Moving from Responsible AI principles to a structured AI management system.
Research library →AI Application Assurance
A practical lens across models, data, RAG, identity, applications and runtime controls.
Research library →Zero Trust for Agentic AI
How machine identities and delegated authority change the traditional Zero Trust model.
Research library →EXPERIENCE BEHIND THE ADVISORY
Devraj Sarkar
Independent AI & Cybersecurity Advisor · C-Suite & Board Advisory
I bring 23+ years of overall IT experience, including 15+ years in consulting, spanning AI strategy, cybersecurity, cloud, DevSecOps, AIOps, governance, risk and enterprise transformation.
I hold an MBA (IT) and ISO/IEC 42001 & ISO/IEC 27001 Lead Auditor credentials. My advisory focus sits at the intersection of enterprise AI, governance, cybersecurity and assurance.
INDEPENDENT AI & CYBERSECURITY ADVISORY
Executive judgment for AI, governance and cyber risk.
Devraj Sarkar is an independent AI and cybersecurity advisor based in Kolkata, India. He advises C-suite and board leaders on AI strategy, Agentic AI, AI governance, AI security, cyber risk and assurance, supporting organizations in India and internationally.
About Devraj Sarkar →01StrategyBusiness-aligned AI
02GovernanceAccountability & controls
03SecurityAI & cyber risk
04AssuranceEvidence & audit
05LeadershipC-suite & board
INDEPENDENT ADVISORY
Complex technology. Clearer risk decisions.
For AI strategy, governance, Agentic AI, security, assurance or cyber-risk requirements in India or internationally, describe the decision you are facing and the context around it.